Search International and National Patent Collections

1. (WO2017155691) DEEP DEFORMATION NETWORK FOR OBJECT LANDMARK LOCALIZATION

Pub. No.:    WO/2017/155691    International Application No.:    PCT/US2017/018839
Publication Date: Fri Sep 15 01:59:59 CEST 2017 International Filing Date: Thu Feb 23 00:59:59 CET 2017
IPC: G06T 7/70
G06T 3/40
G06T 7/50
Applicants: NEC LABORATORIES AMERICA, INC.
Inventors: YU, Xiang
ZHOU, Feng
CHANDRAKER, Manmohan
Title: DEEP DEFORMATION NETWORK FOR OBJECT LANDMARK LOCALIZATION
Abstract:
A system and method are provided. The system includes a processor. The processor is configured to generate a response map for an image, using a four stage convolutional structure. The processor is further configured to generate a plurality of landmark points for the image based on the response map, using a shape basis neural network. The processor is additionally configured to generate an optimal shape for the image based on the plurality of landmark points for the image and the response map, using a point deformation neural network. A recognition system configured to identify the image based on the generated optimal shape to generate a recognition result of the image. The processor is also configured to operate a hardware-based machine based on the recognition result.